David Zhu

Some recommended reading: https://kubernetes.io/docs/concepts/configuration/pod-priority-preemption/#interactions-of-pod-priority-and-qos Takeaways: 1) Set PriorityClass (we do already A+) 2) During resource contention/eviction anyone exceeding their requests (or no request at all) is evicted first. We don't...

/remove-lifecycle stale

/remove-lifecycle stale

according to the [spec](https://github.com/container-storage-interface/spec/blob/master/spec.md): > Max volumes attached | 8 RESOURCE_EXHAUSTED | Indicates that the maximum supported number of volumes that can be attached to the specified node are already...

in fact I would think that it `isFinalError=true` in the `RESOURCE_EXHAUSTED` case

/remove-lifecycle stale

/remove-lifecycle stale

/remove-lifecycle stale /priority backlog

couldn't figure it out before but started work here: https://github.com/kubernetes-sigs/gcp-compute-persistent-disk-csi-driver/pull/189

/remove-lifecycle stale